The present invention relates to a congestion control system of a network and more particularly to a congestion control system in services and applications utilizing the Web.
In the Web service, Web clients of users and Web servers for providing services are connected through a network constructed by a plurality of relay apparatuses to provide the services to the Web clients. When requests are concentrated at the Web servers or the relay apparatuses constructing the network to the extent of exceeding the processing performance thereof, the quality of the whole Web service is degraded even if it is partial congestion.
The Web can be utilized by means of a portable telephone, a portable terminal or the like without temporal and geographical restriction and it is considered that utilization of the Web service is concentrated on specific contents, Web server, time and utilization place in response to a specific time or occurrence of any event.
There is congestion control technique in order to prevent network resources such as processing performance and the like of the Web servers and the relay apparatuses from being exhausted due to the concentration of utilization. The congestion control technique in the Web service is mainly classified into two systems as follows:
A first system is a system in which when requests exceeding an amount of utilizable resources are issued, the requests are cancelled. In this system, control means is provided in a network resource or between the network resource and the clients to relay requests from the clients usually. The control means grasps or monitors an amount of resource to be controlled and an amount of utilization thereof and when the utilization amount exceeds the resource amount or a previously defined value, the control means does not relay requests and cancels the requests.
A second system is a system in which when requests exceeding an amount of utilizable resources are issued, a command is transmitted to the clients issuing the requests to prevent the clients from utilizing the same resource for a fixed time so as to perform connection request regulation. The second system has the same network structure and the congestion judgment method as those of the first system and is different therefrom in that a response for instructing the connection request regulation is returned to the clients instead of cancellation of the requests. The clients which have received the connection request regulation perform the instructed regulation.